us2010discover america in a new centuryThis report has been peer-

reviewed by the Advisory Board of the US2010 Project. Views expressed here are those of the authors.

US2010 ProjectJohn R. Logan, DirectorBrian Stults, Associate Director

Advisory BoardMargo Anderson Suzanne Bianchi Barry Bluestone Sheldon Danziger Claude Fischer Daniel Lichter Kenneth Prewitt

SponsorsRussell Sage FoundationAmerican Communities Project of Brown University

Separate and Unequal: The Neighborhood Gap for Blacks, Hispanics and Asians in Metropolitan AmericaJohn R. LoganBrown UniversityJuly 2011

Summary

The most recent census data show that on average, black and Hispanic households live in neighborhoods with more than one and a half times the poverty rate of neighborhoods where the average non-Hispanic white lives. Even Asians, who have higher incomes than blacks and Hispanics and are less residentially segregated, live in somewhat poorer neighborhoods than whites.

This report analyzes the roots of these disparities and their variation across metropolitan regions. Can they be accounted for by group di�erences in income, because sorting by social class places poorer groups in poorer neighborhoods? There are indeed large income di�erences across racial groups in metropolitan America, but these have little relationship with neighborhood inequality. This report demonstrates that separate translates to unequal even for the most successful black and Hispanic minorities. The average a�uent black or Hispanic household lives in a poorer neighborhood than the average lower-income white resident. Other studies show that neighborhood poverty is associated with inequalities in public schools, safety, environmental quality, and public health. Racial segregation itself is the prime predictor of which metropolitan regions are the ones where minorities live in the least desirable neighborhoods.

1

Introduction This report examines how people’s race/ethnicity and income are translated into racial/ethnic and class segregation across neighborhoods. The first question is to what extent minorities’ residential isolation and limited contact with non-Hispanic whites is a result of income differences. Is segregation a problem of affordability? The second question is how much separate turns out to mean unequal. The study includes all metropolitan regions in the country and examines variations among the regions with the largest minority populations. There are clear results:

• Black household incomes are below 60 percent of non-Hispanic white incomes in the average metropolitan region, while Hispanic household incomes are less than 70 percent. These groups’ relative standing actually became worse between 1990 and 2000 and in the post-2000 years covered by this study. Asians, in contrast, had higher average incomes than non-Hispanic whites in 1990 and they have maintained this advantage over time.

• As black-white segregation has slowly declined since 1990, blacks have become less isolated

from Hispanics and Asians, but their exposure to whites has hardly changed. Affluent blacks have only marginally higher contact with whites than do poor blacks.

• Asians and especially Hispanics have become more isolated from whites as their numbers have

grown, and they both have markedly lower exposure to whites now than they did in 1990. Income is moderately associated with these patterns for Hispanics (that is, affluent Hispanics experience lower isolation and higher contact with whites). Asians’ level of concentration in Asian neighborhoods, however, is unrelated to income, and exposure to whites is only modestly greater for higher-income Asians.

• With only one exception (the most affluent Asians), minorities at every income level live in

poorer neighborhoods than do whites with comparable incomes. Disparities are greatest for the lowest income minorities, and they are much sharper for blacks and Hispanics than for Asians. Affluent blacks and Hispanics live in poorer neighborhoods than whites with working class incomes. There is considerable variation in these patterns across metropolitan regions. But in the 50 metros with the largest black populations, there is none where average black exposure to neighborhood poverty is less than 20 percent higher than that of whites, and only two metros where affluent blacks live in neighborhoods that are less poor than those of the average white.

• The disparity between black and white neighborhood poverty in a metropolitan area is hardly

related to blacks’ average income levels. But racial segregation is a very strong predictor of unequal neighborhoods. Patterns are similar but not as strong for Hispanics. Among Asians, however, parity with whites in neighborhood quality is more closely tied with their own income level.

2

Income disparities and racial segregation We begin with data on the income levels of whites, blacks, Hispanics and Asians in metropolitan regions in each year. Figure 1 (see also Appendix Table 1) presents these data, calculated as the average of a group’s median household income across all metro areas. (Metros are weighted by the number of group households, so that areas with larger populations count more heavily in the average values.) White incomes averaged over $60,000 in the 2005-2009 American Community Survey (ACS), about $25,000 more than blacks and $20,000 more than Hispanics. These are large differences indeed and they are larger now than they were in 1990 in both absolute dollar amounts and the ratio of minority to white income. If people’s neighborhoods simply reflected their incomes, we would expect considerable residential segregation as a result. Asians, on the other hand, have had higher incomes than whites throughout the period. If the affluence of their neighborhoods were commensurate with their income, we would expect them to have very little segregation from whites and in fact to live on average in higher-quality neighborhoods than whites.

These inferences can be directly tested. We measure segregation here with two indicators of the racial composition of the average group member’s neighborhood: the percentage of same group members (referred to in social science research as “isolation”) and the percentage of non-Hispanic whites (“contact”). Table 1 summarizes these indicators in each year for all group households and for households at each of three income levels. These figures tell us the neighborhood characteristic experienced by the average group member across all metropolitan regions in a given year. As noted in the Appendix, the “neighborhood” is defined as the census tract that the household lives in, plus all the surrounding tracts with which it shares a boundary. (For reference, isolation measures are presented for all group households and affluent group households in each of the largest metropolitan regions in Appendix Tables 1-3.)

!

3

Table 1. Racial/ethnic composition of the average household's neighborhood,

by race/ethnicity and household income, in metropolitan regions

Exposure to own group

(isolation) Exposure to whites (contact)

1990 2000 2005-2009 1990 2000 2005-2009

Non-Hispanic white total 83.1 77.9 74.8

Poor 81.7 76.3 74.0

Middle 83.4 78.0 75.1

Affluent 84.0 78.9 75.3

Non-Hispanic black total 47.1 44.4 40.7 41.7 39.0 39.8

Poor 49.6 47.1 42.9 39.7 36.7 38.3

Middle 45.0 43.1 39.7 43.5 40.1 40.5

Affluent 43.1 40.2 36.2 44.6 42.6 42.9

Hispanic total 37.1 40.7 41.8 46.4 40.7 39.5

Poor 41.5 44.8 45.0 41.5 36.4 36.4

Middle 35.7 39.8 41.1 48.2 41.8 40.3

Affluent 29.8 34.4 36.0 54.4 47.1 45.2

Asian total 16.9 16.5 17.5 59.2 53.5 52.1

Poor 16.4 16.5 17.5 56.0 49.2 48.4

Middle 16.7 15.8 16.7 58.4 52.7 51.1

Affluent 17.4 16.8 18.0 62.1 56.7 54.9

Results are provided for whites as a point of comparison. White isolation is very high, partly because whites remain the majority of the population in metropolitan America, but their isolation is declining. The average white household lived in a neighborhood where 83 percent of households were white in 1990. This figure dropped to only 75 percent white in 2005-2009. Note, however, that this outcome is not much related to income. Currently, for example, poor whites live in neighborhoods that average 74.0 percent white, while affluent whites’ neighborhoods average 75.3 percent white. Whites typically live in predominantly white areas, a pattern that is not simply due to their relatively high incomes. Black households are less isolated now than they were in 1990 (40.7 vs. 47.1), but they also now have smaller shares of whites in their neighborhoods (39.8 vs. 41.7). What is changing for them is increasing exposure to Hispanics and Asians, the two fast-growing segments of the population. Affluent blacks are currently less isolated than poor blacks (36.3 vs. 42.9), and also somewhat more exposed to whites (42.9 vs. 39.8). But these differences of three to six points are small in relation to the much larger differences between blacks’ neighborhoods and the metropolitan regions where they live (averaging 63 percent white and only 19 percent black). Race trumps income for blacks.

4

Hispanics are more isolated and have less contact with whites in their neighborhoods in 2005-2009 than in 1990. Their isolation is related to their incomes. In 1990 there was a very large difference in neighborhood racial composition between poor and affluent Hispanics (a difference of about 12 points in isolation and 13 points in contact with whites). These differences persist but they have diminished over time to about 6 points in isolation and 9 points in white contact. However even affluent Hispanics live in neighborhoods that are very different from their metro setting. Their neighborhoods now average 36 percent Hispanic and 45 percent white, but they live in metropolitan areas that where only 25 percent of the population is Hispanic and 55 percent is white. Finally, the pattern for Asians is different from both blacks and Hispanics. Asian isolation increased only slightly despite the rapid Asian population growth, while exposure to whites declined. Asian isolation is unrelated to income, but affluent Asians live in whiter neighborhoods than do poor Asians. Asian exposure to whites is higher than that of blacks or Hispanics. However even affluent Asians live in neighborhoods that are more Asian (18 percent) than the share of Asians in their average metropolitan region (10 percent). In sum, despite very large differences in household income between whites and both blacks and Hispanics, income differences explain only a modest part of the segregation between these groups. Asians have higher incomes than whites, but their higher incomes do not reduce Asians’ isolation. Though higher income does result in greater Asian exposure to whites, even poor Asians experience higher exposure to whites than affluent blacks or Hispanics. Clearly there is a substantial component of segregation that cannot be accounted for by income. Income disparities and exposure to poverty We turn now to the question of the gap in people’s quality of life as measured by the neighborhood’s poverty level. Other studies show that neighborhood poverty is associated with inequalities in public schools, safety, environmental quality, and public health. The US2010 Project’s web pages (http://www.s4.brown.edu/us2010/SeparateAndUnequal/Default.aspx) show in most metro areas similar neighborhood gaps in median and per capita income; percent of residents with a college education or professional occupation; home ownership; and housing vacancy. Data on poverty are used here to illustrate these differences in many other dimensions. Table 2 lists the poverty level of the neighborhood where the average household lived in each year in metropolitan areas across the country. It also evaluates separately the neighborhood environments of poor, middle-income, and affluent group members. We will focus here not on the absolute numbers but on the ratio of the minority group value to the corresponding value for non-Hispanic whites: The higher the ratio, the greater the disparity experienced by the minority group. The overall disparities between groups are generally in line with the differences in their median incomes (as shown in Table 1), and one would be tempted to conclude that blacks and Hispanics live in lower-status neighborhoods than whites and Asians simply because of their own lower earnings. This would be a natural consequence of how a private housing market operates: sorting people by income. Yet it turns out, when we recalculate these figures for households with similar income levels, that racial differences remain large.

5

Table 2. Average neighborhood poverty by race/ethnicity and household income in metropolitan regions

Mean values for group

members Ratio to white mean

1990 2000 2005-2009 1990 2000

2005-2009

Non-Hispanic white total 9.9 9.4 10.7 Poor 12.4 11.6 12.9 Middle 9.9 9.6 10.9 Affluent 7.8 7.7 8.9 Non-Hispanic black total 21.4 18.9 19.0 2.15 2.00 1.77

Poor 24.8 21.9 21.8 2.50 2.32 2.02

Middle 18.8 17.3 17.3 1.89 1.83 1.61

Affluent 15.4 14.3 13.9 1.55 1.52 1.29

Hispanic total 18.9 17.8 17.3 1.90 1.89 1.61

Poor 22.6 20.9 19.9 2.27 2.21 1.85

Middle 17.3 16.8 16.2 1.74 1.78 1.51

Affluent 13.3 13.5 13.0 1.33 1.43 1.21

Asian total 11.7 11.7 11.3 1.18 1.24 1.05

Poor 16.3 16.3 15.2 1.64 1.72 1.41

Middle 11.7 11.9 11.6 1.18 1.26 1.08

Affluent 8.3 8.7 8.7 0.84 0.92 0.81 For example, consider only affluent households whose incomes were above $75,000 in each year (adjusted for inflation). Table 2 shows that the average affluent white household lived in a neighborhood where the poverty share was under 10 percent in every year. But poor white households (incomes below $40,000) lived in neighborhoods with only slightly greater poverty shares, about 12 percent or 13 percent. In contrast, affluent blacks lived in neighborhoods that were 14-15 percent poor, and affluent Hispanics in neighborhoods that were about 13 percent poor. On average around the country, in this whole period of nearly two decades, affluent blacks and Hispanics lived in neighborhoods with fewer resources than did poor whites. Even Asians, whose incomes were higher than whites, lived on average in poorer neighborhoods than whites. The table shows that this disadvantage was mainly due to the residential pattern of poor Asians -- considerably worse than whites of comparable income -- while affluent Asians actually had better outcomes than comparable whites. The level of disparities, as measured by the ratio of minority to white exposure to poverty, declined during the 1990s and continued to drop since 2000. These changes are due to a combination of two factors: a small decline in exposure to poverty for minorities and a small increase – especially since

6

2000 – in exposure to poverty for whites. The latter may reflect the early impacts of the recession and foreclosure crisis that grew more pronounced in 2008 and 2009. Figures 2 and 3 present these disparities in another form that displays the full distribution of people in each group in 2005-2009. In these figures the x-axis measures the neighborhoods’ rank-order position in poverty, from the least poor (the 0th percentile) to the most poor. The y-axis measures the cumulative percentage of group members who live in neighborhoods at that level of poverty or lower. For example, Figure 2 shows that only about 30 percent of black and Hispanic households live in neighborhoods that are at the median (50th percentile) of poverty or less, but more than 60 percent of white and Asian households are in neighborhoods below that level. Figure 2 displays the strong similarity between the distributions for whites and Asians and shows how different is the pattern for blacks and Hispanics.

Figure 3 reproduces the distribution for whites in Figure 2 as a dashed line. It then compares this distribution with that of affluent households (income over $75,000) of each group. This figure shows that even affluent blacks and Hispanics have a poorer neighborhood profile than the average for whites (below the white line on the graph). Affluent whites and Asians, on the other hand, live in markedly better neighborhoods (well above the “Non-Hispanic white total” line).

0  

10  

20  

30  

40  

50  

60  

70  

80  

90  

100  

Percent  of  group  

Percentile  of  neighborhood  poverty  (from  least  to  most  poor)  

Figure  2.    Distribution  of  group  members  by    neighborhood  poverty  level,  2005-­‐2009  

Non-­‐Hispanic  white  

Black  alone  

Hispanic  

Asian  alone  

7

Blacks and whites: a closer look Aggregated national data can mask important regional variations. To begin with, Table 3 shows the average exposure to neighborhood poverty of white, black, and affluent black households in 2005-2009 among the 50 metropolitan regions with the largest black populations. Metros are listed by the size of the disparity as measured by the ratio of black to white exposure.

Table 3. Exposure to poverty for non-Hispanic whites and blacks in 2005-2009 (50 metropolitan regions – MSAs or Metro Divisions – with the most black households in 2005-2009)

White exposure

Black exposure

Affluent black

Ratio black to

white

Ratio affluent black to

white total

1 Newark-Union, NJ-PA MD 5.0 17.9 13.9 3.58 2.79 2 Milwaukee-Waukesha-West Allis, WI MSA 8.5 27.0 20.5 3.17 2.41 3 Philadelphia, PA MD 8.4 24.8 18.4 2.95 2.19 4 Chicago-Joliet-Naperville, IL MD 8.3 22.5 17.9 2.70 2.15 5 Cleveland-Elyria-Mentor, OH MSA 10.2 26.2 18.2 2.55 1.78 6 St. Louis, MO-IL MSA 9.2 22.5 16.9 2.44 1.83 7 Detroit-Livonia-Dearborn, MI MD 13.1 30.6 26.1 2.34 2.00 8 Kansas City, MO-KS MSA 8.9 20.3 14.0 2.29 1.58 9 Minneapolis-St. Paul-Bloomington, MN-WI MSA 8.1 17.7 10.9 2.20 1.35 10 Baltimore-Towson, MD MSA 7.2 14.9 10.4 2.07 1.45 11 Camden, NJ MD 6.4 13.1 8.3 2.05 1.30

0  

10  

20  

30  

40  

50  

60  

70  

80  

90  

100  Percent  of  group  

Percentile  of  neighborhood  poverty  (from  least  to  most  poor)  

Figure  3.    Distribution  of  group  members  by    neighborhood  poverty  level,  2005-­‐    2009:  comparing  white  total  to  afEluent  households  of  each  group  

Non-­‐Hispanic  white  total  

White  -­‐  af@luent  

Black  -­‐  af@luent  

Hispanic  -­‐  af@luent  

Asian  -­‐  af@luent  

8

White exposure

Black exposure

Affluent black

Ratio black to

white

Ratio affluent black to

white total

12 13

Boston-Quincy, MA MD Cincinnati-Middletown, OH-KY-IN MSA

9.0 10.5

18.4 21.2

14.8 15.6

2.05 2.02

1.64 1.49

14 Birmingham-Hoover, AL MSA 10.4 20.9 15.6 2.01 1.50 15 Louisville/Jefferson County, KY-IN MSA 11.5 23.1 15.1 2.00 1.31 16 New York-White Plains-Wayne, NY-NJ MD 10.6 21.2 16.6 1.99 1.56 17 Pittsburgh, PA MSA 11.0 20.8 16.5 1.90 1.51 18 Los Angeles-Long Beach-Glendale, CA MD 10.6 19.4 15.2 1.83 1.44 19 Oakland-Fremont-Hayward, CA MD 8.1 14.9 11.7 1.82 1.44 20 Indianapolis-Carmel, IN MSA 10.2 18.6 13.3 1.82 1.30 21 Washington-Arlington-Alexandria, DC-VA-MD-WV 6.1 11.0 7.9 1.80 1.29 22 Memphis, TN-MS-AR MSA 13.6 24.1 17.5 1.78 1.29 23 West Palm Beach-Boca Raton-Boynton Beach, FL 9.7 17.0 13.6 1.76 1.41 24 Richmond, VA MSA 8.4 14.9 10.5 1.76 1.25 25 Dallas-Plano-Irving, TX MD 10.4 18.2 12.8 1.75 1.24 26 Columbus, OH MSA 11.8 20.7 14.3 1.75 1.21 27 Houston-Sugar Land-Baytown, TX MSA 11.4 19.5 14.7 1.71 1.29 28 Jacksonville, FL MSA 10.2 17.4 13.6 1.70 1.33 29 Nashville-Davidson--Murfreesboro--Franklin, TN 10.9 18.5 13.0 1.70 1.19 30 Tampa-St. Petersburg-Clearwater, FL MSA 11.2 18.5 13.7 1.65 1.22 31 Nassau-Suffolk, NY MD 4.7 7.8 7.3 1.65 1.54 32 Miami-Miami Beach-Kendall, FL MD 13.6 22.3 18.4 1.64 1.35 33 Fort Worth-Arlington, TX MD 10.5 16.9 12.2 1.61 1.17 34 Jackson, MS MSA 13.8 22.0 18.8 1.59 1.37 35 Warren-Troy-Farmington Hills, MI MD 8.2 13.1 10.6 1.58 1.29 36 Atlanta-Sandy Springs-Marietta, GA MSA 9.9 15.3 12.7 1.55 1.28 37 Virginia Beach-Norfolk-Newport News, VA-NC 8.9 13.5 10.7 1.53 1.21 38 New Orleans-Metairie-Kenner, LA MSA 13.5 20.3 17.1 1.51 1.26 39 Baton Rouge, LA MSA 14.6 21.6 17.5 1.49 1.20 40 Charleston-North Charleston-Summerville, SC 12.6 18.2 15.1 1.44 1.20 41 Fort Lauderdale-Pompano Beach-Deerfield Beach, FL 10.9 15.7 12.3 1.44 1.13 42 Charlotte-Gastonia-Rock Hill, NC-SC MSA 10.4 14.8 11.6 1.42 1.11 43 Greensboro-High Point, NC MSA 13.6 18.7 14.0 1.38 1.03 44 Raleigh-Cary, NC MSA 9.6 13.2 10.7 1.37 1.12 45 Augusta-Richmond County, GA-SC MSA 15.2 20.4 16.9 1.34 1.11 46 Las Vegas-Paradise, NV MSA 9.4 12.6 8.9 1.34 0.95 47 Orlando-Kissimmee-Sanford, FL MSA 10.8 14.3 12.0 1.32 1.11 48 Columbia, SC MSA 12.3 15.8 12.7 1.29 1.03 49 Bethesda-Rockville-Frederick, MD MD 4.9 6.1 5.6 1.24 1.14 50 Riverside-San Bernardino-Ontario, CA MSA 11.7 14.4 11.2 1.23 0.96

There are two important regularities in these data. Exposure to poverty is greater among black than other groups in every one of these 50 metros. Even affluent blacks have greater exposure to poverty than the average white in all but two metros (the exceptions are Las Vegas and Riverside).

9

There is also great variation. At one extreme, Newark, black exposure to poverty is three and a half times that of whites (17.9 vs 5.0), and even affluent blacks are in neighborhoods nearly three times as poor as the average white (13.9 vs. 5.0). At the other end of the distribution, blacks’ neighborhoods are only 23 percent poorer than whites’ neighborhoods in Riverside, and affluent blacks’ neighborhoods are slightly less poor. Reviewing the rank order of metros, one notices the familiar pattern associated with residential segregation – large metros mainly in the Northeast and Midwest at one end and mostly Southern and Western metros at the other. What is behind this ordering? The question running through this analysis is whether low average black incomes account for the neighborhood income disparities, or whether segregation by race is the main mechanism. These alternative explanations are probed in Figures 4 and 5. Each figure shows a scatterplot of the level of black neighborhood disadvantage on the y-axis (ratio of black to white exposure to poverty in the metro). Figure 4 arrays metros by black median income on the x-axis, while Figure 5 arrays them by the level of black residential isolation (the percent black in the neighborhood of the average black household). Each figure shows the average regression line and also the R2 (explained variance) for the regression summarizing how strong is the relationship between the two characteristics.

Figure 4 reveals a nearly random pattern, and the R2 is close to zero. Figure 5 shows a much stronger association, with lower disparities in metros with less isolated black populations. The regression line has a clearly positive slope and the explained variance is substantial. These figures suggest that the overall degree of neighborhood disparities for blacks in metropolitan areas are not much related to black income, but are due in large part to residential segregation.

R²  =  0.0197  

0.9  

1.1  

1.3  

1.5  

1.7  

1.9  

2.1  

2.3  

2.5  

2.7  

Ratio  of  black  to  white  exposure  to  poverty  

Black  median  household  income  

Figure  4.  Association  between  black  disadvantage  in  exposure  to  poverty  and  metropolitan  black  median  income,  2005-­‐2009  

10

The Hispanic disadvantage in neighborhood quality Table 4 and Figures 6 and 7 repeat this analysis for Hispanics. Table 4 lists the 50 metro areas with the largest Hispanic populations. As found for blacks, the overall Hispanic-white disparity in exposure to poverty is found in every one of these metros, and in all but three of them, even affluent Hispanics live in poorer neighborhoods than does the average white. There are less extreme values for Hispanics than we saw for blacks (a maximum ratio of about 3.0 in Philadelphia), and there are more cases where affluent Hispanics live in neighborhoods whose poverty level is fairly close to that of the average white.

Table 4. Exposure to poverty for non-Hispanic whites and Hispanics in 2005-2009 (50 metropolitan regions with the most Hispanic households in 2005-2009)

Non-Hispanic

white total

Hispanic total

Affluent Hispanic

Ratio Hispanic to white

Ratio affluent Hispanic to white

total 1 Philadelphia, PA MD 8.4 25.4 13.7 3.02 1.63 2 Hartford-West Hartford-East Hartford, CT MSA 6.8 19.7 12.2 2.91 1.80 3 Newark-Union, NJ-PA MD 5.0 13.3 10.2 2.66 2.03 4 Providence-New Bedford-Fall River, RI-MA MSA 9.7 20.5 16.1 2.11 1.66 5 New York-White Plains-Wayne, NY-NJ MD 10.6 21.6 15.9 2.03 1.50 6 Boston-Quincy, MA MD 9.0 18.0 14.2 2.00 1.58 7 Chicago-Joliet-Naperville, IL MD 8.3 15.3 12.7 1.84 1.52 8 Edison-New Brunswick, NJ MD 5.9 10.3 8.4 1.74 1.42 9 Los Angeles-Long Beach-Glendale, CA MD 10.6 18.2 14.3 1.72 1.36 10 Dallas-Plano-Irving, TX MD 10.4 17.7 13.2 1.70 1.27 11 Phoenix-Mesa-Glendale, AZ MSA 10.9 18.3 14.3 1.69 1.32 12 San Antonio-New Braunfels, TX MSA 11.4 19.0 14.0 1.67 1.23

R²  =  0.2987  

0.9  

1.1  

1.3  

1.5  

1.7  

1.9  

2.1  

2.3  

2.5  

2.7  

0.0   10.0   20.0   30.0   40.0   50.0   60.0  

Ratio  of  black  to  white  exposure  to  povery  

Black  isolation  (%  black  in  the  tract  of  the  average  black  resident)  

Figure  5.    Association  between  black  disadvantage  in  exposure  to  poverty  and  metropolitan  black  isolation,  2005-­‐2009  

11

Non-Hispanic

white total

Hispanic total

Affluent Hispanic

Ratio Hispanic to white

Ratio affluent Hispanic to white

total 13 Houston-Sugar Land-Baytown, TX MSA 11.4 18.6 14.4 1.64 1.27 14 Oxnard-Thousand Oaks-Ventura, CA MSA 7.1 11.7 10.0 1.63 1.40 15 Santa Ana-Anaheim-Irvine, CA MD 7.6 12.3 10.7 1.63 1.42 16 Denver-Aurora-Broomfield, CO MSA 10.0 16.1 11.8 1.61 1.18 17 Fort Worth-Arlington, TX MD 10.5 16.8 12.2 1.60 1.17 18 Tucson, AZ MSA 12.9 20.2 15.4 1.57 1.19 19 Salinas, CA MSA 10.0 15.4 14.2 1.53 1.41 20 Las Vegas-Paradise, NV MSA 9.4 14.3 10.9 1.52 1.16 21 Fresno, CA MSA 16.0 23.9 18.8 1.49 1.18 22 Oakland-Fremont-Hayward, CA MD 8.1 12.1 9.9 1.48 1.21 23 San Diego-Carlsbad-San Marcos, CA MSA 9.9 14.6 11.2 1.47 1.13 24 Austin-Round Rock-San Marcos, TX MSA 11.3 16.3 12.0 1.45 1.07 25 Nassau-Suffolk, NY MD 4.7 6.7 6.1 1.43 1.28 26 Bakersfield-Delano, CA MSA 16.5 23.4 18.9 1.42 1.14 27 San Jose-Sunnyvale-Santa Clara, CA MSA 7.7 10.9 9.8 1.41 1.27 28 West Palm Beach-Boca Raton-Boynton Beach, FL 9.7 13.6 11.4 1.40 1.18 29 Salt Lake City, UT MSA 9.1 12.5 9.6 1.36 1.06 30 Corpus Christi, TX MSA 15.6 20.8 16.8 1.34 1.08 31 Sacramento--Arden-Arcade--Roseville, CA MSA 10.8 14.3 11.6 1.33 1.07 32 Atlanta-Sandy Springs-Marietta, GA MSA 9.9 13.0 11.1 1.31 1.12 33 Albuquerque, NM MSA 12.9 16.7 13.5 1.29 1.05 34 Washington-Arlington-Alexandria, DC-VA-MD-WV MD 6.1 7.8 6.9 1.28 1.12 35 Bethesda-Rockville-Frederick, MD MD 4.9 6.3 5.8 1.28 1.17 36 Stockton, CA MSA 13.5 17.2 14.3 1.28 1.06 37 Modesto, CA MSA 13.3 16.8 14.7 1.26 1.10 38 Tampa-St. Petersburg-Clearwater, FL MSA 11.2 13.9 11.4 1.24 1.01 39 Seattle-Bellevue-Everett, WA MD 9.1 11.2 9.2 1.23 1.02 40 San Francisco-San Mateo-Redwood City, CA MD 8.4 10.3 8.9 1.23 1.06 41 Visalia-Porterville, CA MSA 19.4 23.7 21.0 1.23 1.09 42 El Paso, TX MSA 22.3 27.2 22.5 1.22 1.01 43 Riverside-San Bernardino-Ontario, CA MSA 11.7 14.1 11.7 1.21 1.00 44 Laredo, TX MSA 24.8 29.9 23.3 1.20 0.94 45 Miami-Miami Beach-Kendall, FL MD 13.6 16.3 12.6 1.20 0.93 46 Portland-Vancouver-Hillsboro, OR-WA MSA 11.7 13.6 11.9 1.17 1.02 47 Orlando-Kissimmee-Sanford, FL MSA 10.8 12.0 11.0 1.11 1.01 48 McAllen-Edinburg-Mission, TX MSA 32.8 35.9 33.0 1.10 1.01 49 Brownsville-Harlingen, TX MSA 32.9 35.9 33.4 1.09 1.01 50 Fort Lauderdale-Pompano Beach-Deerfield Beach, FL 10.9 10.5 8.5 0.96 0.78

12

Figures 6 and 7 examine the association between these disparities and the income level and residential isolation of Hispanics. There is little association with Hispanic median income. The association with Hispanic isolation is not as strong as was found for black metropolitan regions, but the explained variance of .08 is still substantial. Evidently, other factors also influence neighborhood disparities for Hispanics. For example, it would be worthwhile to look for effects of immigration status, language assimilation, or education.

R²  =  0.0034  

0.9  

1.1  

1.3  

1.5  

1.7  

1.9  

2.1  

2.3  

2.5  

2.7  

Ratio  of  Hispanic  to  white  exposure  to  poverty  

Hispanic  median  household  income  

Figure  6.    Association  between  Hispanic  disadvantage  in  exposure  to  poverty  and  metropolitan  Hispanic  median  income,  2005-­‐2009  

R²  =  0.0841  

0.9  

1.1  

1.3  

1.5  

1.7  

1.9  

2.1  

2.3  

2.5  

2.7  

0.0   10.0   20.0   30.0   40.0   50.0   60.0   70.0  

Ratio  of  Hispanic  to  white  exposure  to  povery  

Hispanic  isolation  (%  Hispanic  in  the  tract  of  the  average  Hispanic  resident)  

Figure  7.    Association  between  Hispanic  disadvantage  in  exposure  to  poverty  and  metropolitan  Hispanic  isolation,  2005-­‐2009  

13

The Asian advantage: many exceptions Asian Americans are in a very different position overall from blacks and Hispanics. The national averages show that they have higher household incomes than whites, and affluent Asians live in neighborhoods that are less poor than those of comparable whites. Looking more closely at specific metro areas with large Asian populations (we select here the largest 40) reveals many exceptions to these averages. Table 5 shows that Asians live in neighborhoods that are at least 10 percent poorer than the average white in 18 metros, with the greatest disparities (more than 50 percent) in Boston, Philadelphia and Minneapolis. There is a smaller disadvantage in 13 metros. Asians’ neighborhoods are equal to or wealthier than whites’ neighborhoods in nine metros that are widely spread around the country, including Southern California, Florida, the Northeast, Michigan and Texas.

Table 5. Exposure to poverty for non-Hispanic whites and Asians in 2005-2009 (40 metropolitan regions with the most Asian households in 2005-2009)

Non-Hispanic

white total

Asian total

Affluent Asian

Ratio Asian

to white

Ratio affluent Asian to

white total

1 Boston-Quincy, MA MD 9.0 14.6 11.2 1.62 1.25 2 Philadelphia, PA MD 8.4 13.4 8.4 1.59 1.00 3 Minneapolis-St. Paul-Bloomington, MN-WI MSA 8.1 12.2 7.9 1.51 0.97 4 New York-White Plains-Wayne, NY-NJ MD 10.6 13.8 10.9 1.30 1.03 5 Cambridge-Newton-Framingham, MA MD 7.0 8.9 7.5 1.27 1.07 6 Los Angeles-Long Beach-Glendale, CA MD 10.6 13.4 10.8 1.26 1.02 7 Santa Ana-Anaheim-Irvine, CA MD 7.6 9.4 8.2 1.24 1.08 8 Chicago-Joliet-Naperville, IL MD 8.3 10.2 8.4 1.23 1.01 9 Fresno, CA MSA 16.0 19.3 14.1 1.21 0.89 10 Sacramento--Arden-Arcade--Roseville, CA MSA 10.8 13.1 10.3 1.21 0.96 11 Seattle-Bellevue-Everett, WA MD 9.1 10.5 8.6 1.15 0.94 12 Stockton, CA MSA 13.5 15.6 12.9 1.15 0.96 13 San Francisco-San Mateo-Redwood City, CA MD 8.4 9.6 7.8 1.15 0.93 14 Newark-Union, NJ-PA MD 5.0 5.7 4.8 1.15 0.95 15 Oakland-Fremont-Hayward, CA MD 8.1 9.2 7.3 1.13 0.89 16 Houston-Sugar Land-Baytown, TX MSA 11.4 12.7 10.2 1.12 0.89 17 Denver-Aurora-Broomfield, CO MSA 10.0 11.1 8.8 1.11 0.88 18 Fort Worth-Arlington, TX MD 10.5 11.6 8.9 1.11 0.85 19 Baltimore-Towson, MD MSA 7.2 7.9 6.0 1.09 0.83 20 San Diego-Carlsbad-San Marcos, CA MSA 9.9 10.8 8.7 1.09 0.87 21 Honolulu, HI MSA 8.3 9.0 7.5 1.08 0.90 22 Phoenix-Mesa-Glendale, AZ MSA 10.9 11.7 9.4 1.08 0.87 23 Atlanta-Sandy Springs-Marietta, GA MSA 9.9 10.6 8.9 1.06 0.90 24 Vallejo-Fairfield, CA MSA 9.0 9.5 8.5 1.06 0.95 25 Austin-Round Rock-San Marcos, TX MSA 11.3 11.9 8.5 1.06 0.75 26 Tampa-St. Petersburg-Clearwater, FL MSA 11.2 11.8 10.1 1.05 0.90 27 Bethesda-Rockville-Frederick, MD MD 4.9 5.2 4.8 1.05 0.97 28 Columbus, OH MSA 11.8 12.3 7.8 1.04 0.66 29 San Jose-Sunnyvale-Santa Clara, CA MSA 7.7 8.0 7.2 1.04 0.94 30 St. Louis, MO-IL MSA 9.2 9.5 7.6 1.03 0.83 31 Orlando-Kissimmee-Sanford, FL MSA 10.8 10.9 9.8 1.01 0.91

14

32 Portland-Vancouver-Hillsboro, OR-WA MSA 11.7 11.7 9.8 1.00 0.84 33 Las Vegas-Paradise, NV MSA 9.4 9.3 7.8 0.99 0.83 34 Washington-Arlington-Alexandria, DC-VA-MD-WV MD 6.1 6.0 5.4 0.98 0.88 35 Edison-New Brunswick, NJ MD 5.9 5.6 5.1 0.94 0.87 36 Nassau-Suffolk, NY MD 4.7 4.3 4.1 0.92 0.87 37 Dallas-Plano-Irving, TX MD 10.4 9.2 7.3 0.89 0.71 38 Warren-Troy-Farmington Hills, MI MD 8.2 7.3 6.1 0.89 0.74 39 Fort Lauderdale-Pompano Beach-Deerfield Beach, FL 10.9 9.5 8.0 0.87 0.73 40 Riverside-San Bernardino-Ontario, CA MSA 11.7 10.1 8.7 0.86 0.74

What accounts for this variation? Figures 8 and 9 suggest that both income and residential isolation play a role. The clearer pattern is with income: metros where Asians have higher average incomes also tend to place Asians in better neighborhoods relative to whites (R2 = .07). But there is also a small but significant tendency (R2 = .03) for Asians to be in relatively better neighborhoods when they are less residentially isolated. The stronger effect of income may to some extent reflect the different immigration flows of Asians in different parts of the country. Some regions have higher shares of refugees or immigrants from areas like China, which can include both highly qualified professionals and unskilled workers. Another factor is the preference of some Asian groups, including more affluent group members, to live in ethnic communities, which does not necessarily mean living in neighborhoods with fewer resources.

R²  =  0.0669  

0.7  0.8  0.9  1.0  1.1  1.2  1.3  1.4  1.5  1.6  

Ratio  of  Asian  to  white  exposure  to  poverty  

Asian  median  household  income  

Figure  8.    Association  between  Asian  disadvantageor  advantage  in  exposure  to  poverty  and  metropolitan  Asian  median  income,  2005-­‐2009  

15

Separate and Unequal: The Implications In its analysis of the sources of urban riots in the mid-1960s, the National Commission on Civil Disorders observed that the country was dividing into two nations, increasingly separate and unequal. Now – more than four decades later and in a very different social and political climate – new census data remind us that divisions remain very deep. Reductions in black-white segregation have been slow and uneven. New minorities have become much more visible since the 1960s, and while Hispanics and Asians are less segregated than are blacks from whites, their levels of segregation have been unchanged or have increased since 1980. This report provides new information about the racial divide, and reminds us that each group presents a somewhat different profile: 1. The color line for black Americans

Blacks are the most segregated minority and also have the lowest income levels. In the relatively prosperous decade of the 1990s, their incomes nevertheless grew somewhat more slowly than those of whites, so the ratio of black-to-white incomes is lower in 2005-2009 than it was in 1990. Yet the low incomes of blacks are not the main source of either residential segregation or disparities in the resources of the neighborhoods where they live. A central new finding is that blacks’ neighborhoods are separate and unequal not because blacks cannot afford homes in better neighborhoods, but because even when they achieve higher incomes they are unable to translate these into residential mobility.

2. Hispanics in metropolitan America

Hispanics (on average) have a decidedly better position in the American class structure than do blacks, and they live in better neighborhoods, at least as defined by poverty rates. Hispanics

R²  =  0.0321  

0.7  0.8  0.9  1.0  1.1  1.2  1.3  1.4  1.5  1.6  

0.0   2.0   4.0   6.0   8.0   10.0   12.0   14.0   16.0   18.0   20.0  Ratio  of  Asian  to  white  exposure  to  povery  

Asian  isolation  (%  Asian  in  the  tract  of  the  average  Asian  resident)  

Figure  9.    Association  between  Asian  disadvantage  or  advantage  in  exposure  to  poverty  and  metropolitan  Asian  isolation,  2005-­‐2009  

16

nevertheless continue to have lower incomes than whites and live in poorer neighborhoods, even compared to whites with similar incomes. Similar to blacks, their neighborhood gap is not attributable to income differences with whites. The trajectory for Hispanics is negative in two respects. Their incomes are declining relative to whites in both absolute dollar amounts and as a proportion. They also are experiencing growing isolation and declining exposure to non-Hispanic whites. Like blacks, affluent Hispanics live in higher poverty neighborhoods than do whites with working class incomes. Immigration and high fertility rates make Hispanics a rapidly growing population. Changes in their class position and neighborhood attainment probably reflect the characteristics of newer immigrants as much or more than shifts in the fortunes of those who already lived in this country in 1990. But if some Hispanics are indeed experiencing upward mobility, it is not sufficient to counterbalance the disadvantages of the Hispanic population as a whole.

3. The ambiguous standing of Asian Americans

Asians are in the unique position of having higher incomes than whites and, often live in more prosperous neighborhoods. This is not true everywhere: on average, Asians are disadvantaged compared to whites except at the higher-income range. But where there are disparities in neighborhood outcomes, they are relatively modest. In a number of specific metropolitan areas, though, the advantages seen in national averages disappear. In some, we find that Asians have substantially lower incomes than do whites, and in others they enjoy higher income but live in neighborhood of lower quality. The Asian situation, surprisingly, sometimes supports the overall conclusion of this report: Separate in America also means unequal.

This report focused on the neighborhood gap – the differences in characteristics of neighborhoods where black and Hispanic minorities live, compared to whites. Other data not presented here show similar differences in a wide range of other neighborhood measures (median and per-capita income; percent of residents with a college education or professional occupation; home ownership; and housing vacancy), and they confirm that these appear at every income level. We cannot escape the conclusion that more is at work here than simple market processes that place people according to their means. The level of disparities in neighborhood outcomes for blacks and Hispanics has declined in the last 20 years. In this respect, the findings offer some hope for the future, though it would be more encouraging if the change responded more to improvements for minorities than to deterioration for whites. Yet the remaining inequalities are large and surprisingly unaffected by minorities’ incomes. Studies drawing on data from other sources, such as criminal justice, public health, and school statistics, lead to similar conclusions. Separate analyses of school data from the U.S. Department of Education showed that one consequence of school segregation is that minority children are enrolled in schools with much higher levels of poverty, as indicated by eligibility for reduced-price school lunches. The average black or Hispanic child in a public elementary school in 1999-2000 was in a school where more than 65 percent of students were poor. This compared to 42 percent poor in the average Asian child’s school, but only 30 percent poor in the average white child’s school.

17

Residential segregation is not benign. It does not mean only that blacks and Hispanics, Asians and whites live in different neighborhoods with little contact between them. It means that whatever their personal circumstances, black and Hispanic families on average live at a disadvantage and raise their children in communities with fewer resources.

18

Appendix: Data Sources and Methodology These analyses are based on census tract data from the Census of Population 1990 (STF4A), the Census of Population 2000 (SF3), and the 2005-2009 American Community Survey (ACS). These sources include tables listing the household income distribution for specific racial and ethnic groups in every tract. All income data referred to in this report are for households, classified by the race/ethnicity of the household head. Income data for 1990 and 2000 have been adjusted to 2009 dollars, which is also the reference for income data reported in the ACS. To accomplish this, the 2000 income figures were multiplied by 1.2459; 1990 income figures were multiplied by 1.6171. We aggregated data from census tracts in each year to provide totals for metropolitan regions, using the official 2010 boundaries of metropolitan regions. Income data for all three years are taken directly from tables prepared by the Census Bureau for non-Hispanic whites (people who reported only white race) and Hispanics. It would be preferable to identify a non-Hispanic black category of persons who reported their race as black alone or in combination with another race. Because this is not possible for all three years, we instead define “black” households as those headed by persons who reported only black race, without regard to Hispanic origin. The same approach is used to identify Asians. For convenience, we use the terms white (or non-Hispanic white), black, Hispanic and Asian to refer to these groups. Median incomes have been estimated from the grouped income data. To facilitate a breakdown of residential patterns by the income level of households, incomes have been categorized into three consistent categories: "poor" (income below 175 percent of the poverty line for a family of four, "affluent" (income more than 350 percent of the poverty line,), and "middle income" (those falling in between). Our choices of cutting points were constrained by the categories provided in the data. For “poor” we used values under $22,500 in 1990, $30,000 in 2000, and $40,000 in 2005-2009. For “affluent” we used values over $45,000 in 1990, $60,000 in 2000, and $75,000 in 2005-2009. In the following tables, neighborhood quality is measured as the percentage of families below the official poverty line. The ACS calculates these data taking into account both size and age composition of families. The US2010 website (www.s4.brown.edu/us2010) provides similar tabulations for a variety of other neighborhood characteristics: median income, per capita income, education level, occupation, homeownership, housing vacancy, US-born share of the population, and share of recent immigrants. The figures are exposure indices: they show the values for the neighborhood where the average group household lives. In addition two segregation measures are used here: isolation (the share of same-group population in the neighborhood where the average group member lives) and exposure to whites (the share of non-Hispanic white population in the neighborhood where the average group member lives). Typically researchers use characteristics of the census tract where people live as a measure of their “neighborhood.” In this report we use a larger area: the census tract plus each adjacent tract. There are several advantages of this approach which is now possible through computer mapping techniques. First, many studies have shown that people are affected not only by conditions in their own tract but also by the larger area in which the tract is embedded. These are often referred to as “spatial” effects. Second, especially for people who live near a tract’s outer edge, residents often live in closer proximity to many people in an adjacent tract than to many people in their own, and it makes sense to take the adjacent tract into account. Third, there are potential problems with the reliability of data from a single tract, especially for the socioeconomic characteristics on which we focus. The original source for

19

information in 1990 and 2000 is the Census long form questionnaire, which was completed by only a 1-in-6 sample of households. The 2005-2009 American Community Survey data are based on even smaller samples. Furthermore, a substantial share of Americans in each year provides no answers to key questions such as income, and the Census Bureau filled in the missing information with imputed data for households that were similar in other respects. Hence all of these estimates are affected by both missing data and sampling error. Dealing with groups of adjacent tracts rather than single tracts should improve the reliability of data.

Appendix Table 1. Median household income in metropolitan regions by race and Hispanic origin (adjusted to 2009 dollars)

1990 2000 2005-2009

Non-Hispanic white $56,240 $62,066 $61,706

Black $34,496 $39,056 $37,047 Black ratio to white 0.613 0.629 0.600 White-black difference $21,745 $23,010 $24,659

Hispanic $40,566 $43,509 $42,098 Hispanic ratio to white 0.721 0.701 0.682 White-Hispanic difference $15,675 $18,557 $19,609

Asian $60,974 $66,822 $70,568 Asian ratio to white 1.084 1.077 1.144 White-Asian difference -$4,734 -$4,756 -$8,862

20

Appendix Table 2. Trends in black household isolation, all households and affluent households (50 metropolitan regions with the most black households in 2005-2009)

Isolation: All households

Isolation: Affluent households

2005-09 2000 1990 2005-09 2000 1990

1 Detroit-Livonia-Dearborn, MI MD 76.4 79.6 80.0 73.2 77.4 78.1 2 Memphis, TN-MS-AR MSA 63.0 65.8 66.2 55.6 59.1 61.9 3 Cleveland-Elyria-Mentor, OH MSA 62.8 66.4 70.2 52.6 57.7 59.2 4 Chicago-Joliet-Naperville, IL MD 62.2 65.4 70.0 56.6 59.8 64.5 5 Jackson, MS MSA 59.6 60.9 60.8 54.8 58.1 58.6 6 Philadelphia, PA MD 59.5 62.7 66.7 49.5 55.1 60.2 7 Milwaukee-Waukesha-West Allis, WI MSA 58.9 61.4 63.7 50.6 54.8 53.4 8 Birmingham-Hoover, AL MSA 57.6 62.2 61.3 48.0 52.8 53.4 9 Washington-Arlington-Alexandria, DC-VA-MD-WV MD 57.6 60.8 63.1 53.8 56.2 58.1 10 St. Louis, MO-IL MSA 56.9 59.0 60.7 47.4 51.2 49.4 11 Baltimore-Towson, MD MSA 56.6 59.3 61.3 47.0 50.5 52.5 12 Newark-Union, NJ-PA MD 56.4 60.0 62.8 49.3 52.7 55.4 13 New Orleans-Metairie-Kenner, LA MSA 53.2 62.2 59.8 48.2 58.6 53.9 14 Atlanta-Sandy Springs-Marietta, GA MSA 53.0 56.6 57.1 50.4 53.6 55.9 15 Baton Rouge, LA MSA 52.9 53.3 51.1 45.7 48.4 46.9 16 New York-White Plains-Wayne, NY-NJ MD 50.0 54.7 56.0 50.9 53.8 54.7 17 Miami-Miami Beach-Kendall, FL MD 46.8 49.1 49.3 43.0 47.5 45.4 18 Columbia, SC MSA 45.8 47.1 46.8 42.4 44.2 45.8 19 Richmond, VA MSA 45.1 49.4 50.9 36.3 40.8 42.0 20 Augusta-Richmond County, GA-SC MSA 43.8 44.8 43.2 38.8 41.3 37.5 21 Virginia Beach-Norfolk-Newport News, VA-NC MSA 43.4 45.2 45.6 37.8 39.0 37.9 22 Jacksonville, FL MSA 42.5 45.3 48.0 35.4 38.8 43.1 23 Kansas City, MO-KS MSA 41.8 48.3 53.8 29.6 37.2 40.7 24 Fort Lauderdale-Pompano Beach-Deerfield Beach, FL MD 41.4 39.4 36.1 34.6 33.6 30.8 25 Indianapolis-Carmel, IN MSA 41.1 47.7 52.0 32.7 42.0 46.3 26 Louisville/Jefferson County, KY-IN MSA 39.3 42.9 46.1 26.5 32.7 37.0 27 Cincinnati-Middletown, OH-KY-IN MSA 38.6 42.6 43.6 32.4 34.4 32.6 28 Greensboro-High Point, NC MSA 38.1 40.1 41.1 31.5 34.4 37.5 29 Charleston-North Charleston-Summerville, SC MSA 37.9 41.3 44.8 32.5 36.1 39.3 30 Columbus, OH MSA 36.9 38.9 41.5 28.6 31.0 33.1 31 Charlotte-Gastonia-Rock Hill, NC-SC MSA 34.8 37.6 39.8 28.7 32.7 34.3 32 Nashville-Davidson--Murfreesboro--Franklin, TN MSA 34.6 36.7 40.3 26.0 30.3 35.4 33 Boston-Quincy, MA MD 34.4 38.8 46.2 32.2 35.3 44.4 34 West Palm Beach-Boca Raton-Boynton Beach, FL MD 33.6 36.4 42.3 25.2 30.4 36.4 35 Warren-Troy-Farmington Hills, MI MD 32.6 36.3 34.6 32.0 34.4 30.8 36 Dallas-Plano-Irving, TX MD 32.1 34.8 40.9 27.9 29.6 38.2 37 Pittsburgh, PA MSA 31.7 35.6 37.3 26.3 29.7 30.1 38 Houston-Sugar Land-Baytown, TX MSA 31.7 37.4 42.3 27.0 32.5 38.8 39 Tampa-St. Petersburg-Clearwater, FL MSA 30.0 33.6 36.8 21.6 26.6 27.8 40 Raleigh-Cary, NC MSA 29.0 30.6 33.9 26.2 26.6 30.1 41 Orlando-Kissimmee-Sanford, FL MSA 28.3 28.9 28.5 21.2 23.5 25.1 42 Nassau-Suffolk, NY MD 28.1 28.8 28.5 28.8 29.3 29.1 43 Los Angeles-Long Beach-Glendale, CA MD 27.7 31.3 38.9 27.0 30.9 36.9 44 Camden, NJ MD 27.5 28.1 29.1 24.2 26.0 25.0 45 Fort Worth-Arlington, TX MD 24.4 28.3 34.1 21.3 22.9 27.8 46 Oakland-Fremont-Hayward, CA MD 23.3 29.9 40.1 19.0 23.9 32.2 47 Bethesda-Rockville-Frederick, MD MD 22.1 21.9 19.2 21.1 20.7 17.9 48 Minneapolis-St. Paul-Bloomington, MN-WI MSA 16.4 17.1 18.6 11.7 13.6 12.9 49 Las Vegas-Paradise, NV MSA 13.3 15.1 23.7 12.1 12.7 19.7 50 Riverside-San Bernardino-Ontario, CA MSA 9.9 11.4 10.1 9.2 10.7 9.6

21

Appendix Table 3. Trends in Hispanic household isolation, all households and affluent households (50 metropolitan regions with the most Hispanic households in 2005-2009)

Isolation: All households

Isolation: Affluent households

2005-09 2000 1990 2005-09 2000 1990

1 Laredo, TX MSA 94.1 94.2 93.9 92.3 92.0 91.5 2 McAllen-Edinburg-Mission, TX MSA 89.4 88.3 85.4 87.5 85.3 81.9 3 Brownsville-Harlingen, TX MSA 86.6 85.4 82.9 85.0 83.8 80.4 4 El Paso, TX MSA 82.4 80.0 72.9 79.6 76.0 65.6 5 Miami-Miami Beach-Kendall, FL MD 71.6 68.5 64.4 68.8 66.3 61.2 6 Salinas, CA MSA 65.5 60.8 47.4 61.9 58.2 42.6 7 Corpus Christi, TX MSA 63.1 61.7 60.2 57.1 55.5 51.4 8 San Antonio-New Braunfels, TX MSA 62.4 62.3 60.7 51.6 52.7 48.6 9 Los Angeles-Long Beach-Glendale, CA MD 60.7 58.6 52.7 54.7 53.1 46.3 10 Visalia-Porterville, CA MSA 59.1 53.6 41.1 54.4 49.3 37.1 11 Bakersfield-Delano, CA MSA 56.5 50.6 42.1 48.2 41.5 30.7 12 Fresno, CA MSA 54.5 50.8 42.1 46.8 44.4 35.5 13 Albuquerque, NM MSA 51.2 49.4 45.3 46.7 45.3 40.2 14 Oxnard-Thousand Oaks-Ventura, CA MSA 51.0 47.8 38.6 44.4 43.4 33.8 15 Riverside-San Bernardino-Ontario, CA MSA 50.1 44.7 31.9 46.8 41.5 29.3 16 Santa Ana-Anaheim-Irvine, CA MD 47.6 45.7 36.7 41.7 39.4 30.9 17 Tucson, AZ MSA 45.4 44.6 40.2 37.7 36.4 32.0 18 Houston-Sugar Land-Baytown, TX MSA 45.3 42.4 33.4 37.0 35.4 26.1 19 Phoenix-Mesa-Glendale, AZ MSA 44.1 39.9 28.9 36.2 31.5 21.5 20 Modesto, CA MSA 43.4 36.3 25.3 40.0 34.1 23.2 21 New York-White Plains-Wayne, NY-NJ MD 42.9 42.9 39.5 35.0 35.1 31.0 22 San Diego-Carlsbad-San Marcos, CA MSA 42.4 39.0 30.3 35.9 32.4 24.3 23 Dallas-Plano-Irving, TX MD 41.5 37.3 25.0 32.2 30.6 19.4 24 Chicago-Joliet-Naperville, IL MD 41.0 41.5 35.7 34.5 35.0 26.6 25 Stockton, CA MSA 40.2 35.1 27.7 36.4 31.0 23.8 26 Austin-Round Rock-San Marcos, TX MSA 38.7 35.0 27.8 31.6 29.5 21.7 27 San Jose-Sunnyvale-Santa Clara, CA MSA 37.2 35.7 30.7 35.0 33.4 27.5 28 Las Vegas-Paradise, NV MSA 37.2 30.6 13.5 29.9 25.3 11.2 29 Denver-Aurora-Broomfield, CO MSA 35.3 33.4 26.4 27.5 26.6 17.7 30 Fort Worth-Arlington, TX MD 34.2 30.4 21.8 25.6 24.8 15.4 31 Newark-Union, NJ-PA MD 34.1 31.6 28.1 27.3 25.3 21.1 32 Orlando-Kissimmee-Sanford, FL MSA 30.8 23.4 10.8 27.8 21.6 10.1 33 Oakland-Fremont-Hayward, CA MD 29.4 24.8 16.8 25.4 22.1 15.4 34 Hartford-West Hartford-East Hartford, CT MSA 29.1 29.8 26.5 18.5 18.3 14.5 35 Providence-New Bedford-Fall River, RI-MA MSA 28.2 24.2 12.4 22.2 20.1 9.5 36 Fort Lauderdale-Pompano Beach-Deerfield Beach, FL MD 28.1 20.5 10.2 31.0 22.7 10.7 37 Edison-New Brunswick, NJ MD 26.0 22.7 17.3 20.5 17.9 12.7 38 San Francisco-San Mateo-Redwood City, CA MD 25.9 26.1 22.7 22.8 24.7 20.6 39 West Palm Beach-Boca Raton-Boynton Beach, FL MD 25.4 19.2 12.0 21.0 16.0 10.4 40 Philadelphia, PA MD 24.2 23.5 21.7 10.4 10.4 7.8 41 Salt Lake City, UT MSA 23.6 18.5 9.2 19.8 16.1 7.3 42 Sacramento--Arden-Arcade--Roseville, CA MSA 23.4 20.3 15.2 20.3 17.6 13.5 43 Tampa-St. Petersburg-Clearwater, FL MSA 22.5 19.2 14.4 19.5 17.2 12.4 44 Boston-Quincy, MA MD 22.2 20.1 14.5 19.4 18.6 12.6 45 Nassau-Suffolk, NY MD 21.1 17.8 10.9 19.6 17.2 10.4 46 Bethesda-Rockville-Frederick, MD MD 19.1 15.0 9.2 17.2 13.6 7.8 47 Washington-Arlington-Alexandria, DC-VA-MD-WV MD 18.3 16.3 10.7 16.4 14.2 8.5 48 Atlanta-Sandy Springs-Marietta, GA MSA 17.5 13.0 3.1 13.7 12.0 2.7 49 Portland-Vancouver-Hillsboro, OR-WA MSA 13.5 10.0 4.2 11.6 9.1 3.6 50 Seattle-Bellevue-Everett, WA MD 9.7 6.4 3.0 8.2 5.8 2.8

22

Appendix Table 4. Trends in Asian household isolation, all households and affluent households

(50 metropolitan regions with the most Asian households in 2005-2009)

Isolation: All households

Isolation: Affluent households

2005-09 2000 1990

2005-09 2000 1990

1 Honolulu, HI MSA 55.2 52.4 68.0 53.9 51.3 67.6 2 San Jose-Sunnyvale-Santa Clara, CA MSA 38.1 32.3 21.6 39.4 33.3 22.1 3 San Francisco-San Mateo-Redwood City, CA MD 36.6 35.0 33.1 35.1 33.7 30.9 4 Oakland-Fremont-Hayward, CA MD 29.0 24.5 18.6 30.8 25.5 18.4 5 Los Angeles-Long Beach-Glendale, CA MD 26.4 23.9 19.5 25.6 23.4 18.7 6 Santa Ana-Anaheim-Irvine, CA MD 25.0 21.2 14.2 24.3 19.8 13.7 7 New York-White Plains-Wayne, NY-NJ MD 24.2 21.1 16.6 20.3 17.5 12.9 8 Vallejo-Fairfield, CA MSA 20.1 17.8 17.1 21.2 19.0 18.6 9 Edison-New Brunswick, NJ MD 19.9 15.5 6.9 20.5 15.9 6.7 10 Stockton, CA MSA 18.4 16.9 20.4 18.0 14.9 17.5 11 San Diego-Carlsbad-San Marcos, CA MSA 18.3 16.8 12.9 19.5 18.3 13.5 12 Sacramento--Arden-Arcade--Roseville, CA MSA 17.3 14.7 13.3 17.0 14.4 13.8 13 Seattle-Bellevue-Everett, WA MD 16.9 14.7 13.7 16.5 13.8 11.6 14 Bethesda-Rockville-Frederick, MD MD 14.7 12.4 8.9 15.2 12.8 9.1 15 Washington-Arlington-Alexandria, DC-VA-MD-WV MD 13.3 10.5 7.2 13.7 10.7 7.3 16 Boston-Quincy, MA MD 12.9 12.5 10.7 11.1 9.9 7.0 17 Chicago-Joliet-Naperville, IL MD 12.8 11.4 9.6 11.5 10.3 7.9 18 Houston-Sugar Land-Baytown, TX MSA 11.7 10.7 7.7 12.1 10.9 8.0 19 Fresno, CA MSA 11.2 9.7 11.5 10.5 8.5 7.4 20 Dallas-Plano-Irving, TX MD 11.1 8.1 4.6 12.1 8.5 4.6 21 Cambridge-Newton-Framingham, MA MD 11.0 9.1 6.0 10.2 7.8 4.3 22 Las Vegas-Paradise, NV MSA 10.2 6.7 4.0 10.7 6.8 3.8 23 Riverside-San Bernardino-Ontario, CA MSA 9.5 8.2 5.6 10.5 9.2 6.1 24 Newark-Union, NJ-PA MD 9.3 7.1 4.4 9.5 7.5 4.5 25 Nassau-Suffolk, NY MD 9.1 6.2 3.7 9.6 6.6 3.9 26 Warren-Troy-Farmington Hills, MI MD 8.9 6.4 3.4 9.7 7.0 3.9 27 Portland-Vancouver-Hillsboro, OR-WA MSA 8.6 7.0 4.9 9.3 7.4 4.6 28 Minneapolis-St. Paul-Bloomington, MN-WI MSA 8.5 8.3 6.4 7.1 6.1 3.0 29 Philadelphia, PA MD 8.4 6.8 4.8 7.1 5.6 3.5 30 Atlanta-Sandy Springs-Marietta, GA MSA 8.4 6.3 3.6 8.7 6.2 3.1 31 Austin-Round Rock-San Marcos, TX MSA 7.2 6.0 4.2 7.4 5.6 2.8 32 Baltimore-Towson, MD MSA 6.7 5.0 3.2 7.2 5.2 3.4 33 Fort Worth-Arlington, TX MD 6.3 5.1 3.8 6.1 5.0 3.2 34 Columbus, OH MSA 5.9 5.0 3.4 5.9 4.6 3.0 35 Orlando-Kissimmee-Sanford, FL MSA 4.9 3.6 2.4 5.1 3.8 2.5 36 Denver-Aurora-Broomfield, CO MSA 4.4 3.7 2.7 4.7 3.7 2.5 37 Phoenix-Mesa-Glendale, AZ MSA 4.3 3.2 2.4 4.4 3.2 2.1 38 St. Louis, MO-IL MSA 4.1 3.2 1.9 4.5 3.2 2.1 39 Tampa-St. Petersburg-Clearwater, FL MSA 4.1 2.8 1.7 4.0 2.8 1.5 40 Fort Lauderdale-Pompano Beach-Deerfield Beach, FL MD 3.8 2.8 1.5 4.2 3.0 1.6